The incidence of prostate cancer is rapidly overcoming all other cancers in men 50 years and older and is now the second leading cause of cancer deaths in men. Prostate cancer can be broadly classified into three forms: latent, aggressive and metastatic. Each of these forms can be pathologically "staged" with respect to their anatomical location in or beyond the prostate and their degree of differentiation (Gleason's grade). Since a tumor results from a series of chromosomal alterations which allow the cell to escape from the normal mechanisms which control its growth, then it would be a logical first step to identify the specific chromosomal changes which are associated with tumor development. However, a paucity of information exists on the genetic and molecular evolution of events which are responsible for prostate cancer. To date, most attempts at identifying karyotypic chromosomal rearrangements in prostate cancer biopsies have proven to be frustrating. In part this is due to the cellular growth characteristics of prostate cancers which can lie dormant for months before division. In addition to their slow growth rate, obtaining metaphases from prostate cancer biopsies also has the inherent problem of selection for rapidly growing cells, biasing the results to cells that may not be representative of the tumor. Therefore, to avoid the problems associated with conventional cytogenetic banding techniques, we intend to apply the techniques of fluorescent in situ hybridization combine with premature chromosome condensation to karyotype prostate cancer cells from biopsies. With this new approach, the need to grow cells for even short periods of time can be avoided, and the population of cells to be analyzed is only limited by the size of the biopsy. this will allow direct analysis of tumor cells in situ, a goal that is difficult to achieve by conventional cytogenetic analysis because of the requirement for cell cultures to obtain metaphase chromosomes for banding analysis. We will use chromosome specific DNA libraries as probes to detect gross structural aberrations for each human chromosome; chromosome specific repetitive probes such as alpha satellite DNA (centromere specific probes) to detect numerical chromosome changes; and cosmid or YAC (Yeast Artificial Chromosomes) probes specific for microchromosomal regions that have putatively been implicated in prostate cancer such as 7q24 and 10q24 as well as those we will find during the course of this study. The ultimate goal of this study is to identify chromosome alterations which are stage or differentiation specific for prostate cancer. This goal has not yet been achieved for prostate cancer, mainly due to the problems of obtaining sufficient material for cytogenetic analysis. In addition, knowledge of known cytogenetic changes in prostate cancers will be useful both for prognosis and detection of minimal residual disease.